โจ Inspiration
Let's be real, trying to figure out courses at UCR (and tbh, most unis ๐ฌ) is kinda a hot mess ๐ฅ. Info is scattered everywhere, some departments are gatekeeping ๐คซ, and the stuff you do find is confusing af. Trying to plan your 4-year sched? Especially as a freshman? Good luck, fam ๐ฉ. We were tired of the stress, so we decided to build something to fix it. โ๏ธ It's the disorganization and the struggle to plan that really got us wanting to create Classify.
๐ What it does
Meet Classify! โจ Your new academic BFF. It's got this super smart AI Chatbot ๐ค โ just tell it what courses you've slayed so far, and it'll recommend what's next on your academic glow-up journey. But wait, there's more! ๐ It doesn't just list courses; it maps them out in a visually stunning โจ prerequisite flowchart. It's interactive too! Click on any course bubble to get the lowdown (course description, units, etc.). Planning just got a major level up. ๐
๐ ๏ธ How we built it
We cooked this up with some cool tech! ๐งโ๐ป Here's the stack:
- Next.js โก๏ธ (Frontend & Backend vibes)
- React โ๏ธ (Making the UI pop)
- TypeScript โจ๏ธ (Keeping our code clean)
- Tailwind CSS ๐ฌ๏ธ (For that โจaestheticโจ styling)
- shadcn/ui ๐จ (Sweet pre-built components)
- Google Gemini AI ๐ง โจ (The brains behind the chatbot)
- React Flow ๐ (Making those flowcharts flow)
๐ญ Challenges we ran into
Ngl, the struggle was real sometimes ๐ .
- Data Hunt: Hunting down course data scattered across random UCR websites felt like a treasure hunt gone wrong ๐บ๏ธโ. Seriously, why is it hidden like that?!
- Schema Drama: Figuring out how to organize all that inconsistent data (the schema) gave us major headaches ๐คฏ.
- Flowchart Boss Battle: Don't even get us started on making the prerequisite flowchart algorithm actually work and display everything correctly โ that was a whole boss battle! ๐พ Getting those lines and prereqs to look good and be accurate took some serious brainpower.
๐ Accomplishments that we're proud of
Honestly, we kinda slayed this? ๐ We're super proud of:
- Building a fully functional full-stack web app ๐ป that actually works!
- Doing it with just two people in literally 24 hours ๐คฏ. The grind was intense!
- Surviving on basically zero sleep โ๏ธ๐ซ๐ค.
- Still managing to have a blast doing hackathon side quests like karaoke ๐ค๐. It's about the journey, right?
- Mad props to the team! ๐
๐ง What we learned
This hackathon was a crash course! ๐ข We definitely:
- Leveled up our full-stack skills (shoutout Next.js!).
- Learned a lot about wrangling messy, real-world data ๐งน (turns out uni websites aren't perfect APIs lol).
- Flexed our algorithm design muscles ๐ช making that flowchart logic happen.
- Mastered the art of teamwork under extreme pressure and caffeine fumes. Legendary skills unlocked. โจ
- Confirmed that karaoke breaks are, in fact, essential for survival. ๐
๐ฎ What's next for Classify
Classify is just getting started! ๐คฉ We've got big plans:
- Sentiment Analysis: Adding personalization by checking course sentiment (using RAG on Rate My Professor & Reddit data ๐) โ should you really take that class? ๐ค
- RAG Implementation: Diving deeper into Retrieval-Augmented Generation for even smarter, context-aware chat responses.
- Viz Glow-Up: Making the flowchart visualization even more โจaestheticโจ and customizable.
- Sharing is Caring: Letting users share their dope schedules with friends ๐ค.
- World Domination (jk... kinda): Expanding Classify to support more universities! ๐ Get ready. ๐
Built With
- gemini
- nextjs
- react-flow
- tailwind
- typescript

Log in or sign up for Devpost to join the conversation.